JLNTV_IPD/CogVideo
4
0
Fork 0
mirror of https://github.com/THUDM/CogVideo.git synced 2025-04-06 03:57:56 +08:00
Code Issues Packages Projects Releases Wiki Activity

Merge remote-tracking branch 'upstream/main' into dev

Browse Source
This commit is contained in:
OleehyO 2024-12-16 11:38:26 +00:00
commit cfaca91cde
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
finetune/train_cogvideox_image_to_video_lora.py
View File

@ -1246,11 +1246,11 @@ def main(args):
use_deepspeed_optimizer = ( use_deepspeed_optimizer = (
accelerator.state.deepspeed_plugin is not None accelerator.state.deepspeed_plugin is not None
and accelerator.state.deepspeed_plugin.deepspeed_config.get("optimizer", "none").lower() == "none" and accelerator.state.deepspeed_plugin.deepspeed_config.get("optimizer", "none").lower() != "none"
) )
use_deepspeed_scheduler = ( use_deepspeed_scheduler = (
accelerator.state.deepspeed_plugin is not None accelerator.state.deepspeed_plugin is not None
and accelerator.state.deepspeed_plugin.deepspeed_config.get("scheduler", "none").lower() == "none" and accelerator.state.deepspeed_plugin.deepspeed_config.get("scheduler", "none").lower() != "none"
) )
optimizer = get_optimizer(args, params_to_optimize, use_deepspeed=use_deepspeed_optimizer) optimizer = get_optimizer(args, params_to_optimize, use_deepspeed=use_deepspeed_optimizer)
@ -1283,7 +1283,7 @@ def main(args):
image_noise_sigma = torch.normal(mean=-3.0, std=0.5, size=(1,), device=image.device) image_noise_sigma = torch.normal(mean=-3.0, std=0.5, size=(1,), device=image.device)
image_noise_sigma = torch.exp(image_noise_sigma).to(dtype=image.dtype) image_noise_sigma = torch.exp(image_noise_sigma).to(dtype=image.dtype)
noisy_image = torch.randn_like(image) * image_noise_sigma[:, None, None, None, None] noisy_image = image + torch.randn_like(image) * image_noise_sigma[:, None, None, None, None]
image_latent_dist = vae.encode(noisy_image).latent_dist image_latent_dist = vae.encode(noisy_image).latent_dist
return latent_dist, image_latent_dist return latent_dist, image_latent_dist